II.

4 Project Results: Advanced Combustion

II.4.4 Characterization of Coal and Biomass Conversion Behaviors in

Advanced Energy Systems

Investigators
Reginald E. Mitchell, Associate Professor, Mechanical Engineering; Paul A. Campbell
and Liqiang Ma, Graduate Researchers; Lars Sørum, Visiting Scholar
Introduction
The goal of this project is to develop models that predict accurately coal and biomass
gasification and combustion behaviors in the type environments likely to be established
in advanced energy systems. This requires acquiring the information needed to
understand and characterize the fundamental chemical and physical processes that govern
coal and biomass conversion at high temperatures and pressures. The models can be used
to determine operating conditions that optimize thermal efficiency and to examine design
strategies for integrating combined cycles for the production of synthesis gas and electric
power with minimum impact on the environment.
There is considerable concern regarding the potential global environmental impact of
fossil fuels used for power generation. Carbon dioxide emissions are among the
concerns. Coal will play a significant role in meeting the world’s energy demands for the
next fifty to one hundred years even if hydrogen becomes the primary energy carrier. By
increasing the fraction of renewable energy in the energy supply, the extent to which
carbon dioxide emissions will impact atmospheric properties can be mitigated. Biomass
is a renewable fuel, and is considered as being CO2-neutral with respect to the
greenhouse gas balance if the use of fossil fuels in harvesting and transporting the
biomass is not considered. Co-firing biomass with coal in traditional coal-fired boilers
and furnaces or using biomass-derived gas as a reburn fuel in coal-fired systems represent
two options for combined renewable and fossil energy utilization. Configurations that
employ both biomass and coal in integrated gasification, combined gas and steam power
cycles and hybrid technologies that produce synthesis gas for fuel cells as well as produce
electric power in combined gas and steam power cycles offer additional options.
Presently, the United States Department of Energy’s Office of Fossil Energy is
considering developing hybrid gasifier-combustor energy systems as the core technology
for the Department’s Vision 21 energy plant of the future [1]. Hybrid technologies are
well suited for combined renewable/fossil energy utilization. A number of potential
approaches to the co-utilization of coal and biomass have been the subject of
demonstration projects, both in Europe and the United States [2]. The selection of any
particular co-firing option is likely to be made on the basis of minimum interference with
normal operation of the coal-fired facility with minimum impact on its environmental
control equipment. The design of efficient coal/biomass co-utilization energy systems
with integrated thermal management to minimize waste heat requires an understanding of
the processes that control the physical transformations that fuel particles undergo when
exposed to hot environments and the chemical reactions responsible for conversion of the
solid material to gaseous species and ash. The goal of this project is to provide the
needed understanding. Our efforts will result in fundamentals-based sub-models for
particle mass loss, size, apparent density, and specific surface area evolution during

202 GCEP Technical Report - 2004

 II.4 Project Results: Advanced Combustion

conversion of coal and biomass material to gas-phase species during gasification and
combustion processes.

Background
In the high-temperature oxidizing environments established in coal-fired boilers and
furnaces, the char particles formed subsequent to coal devolatilization burn with
decreases in both size and apparent density, and the relative changes in size and density
depend upon the extent of char conversion. To capture this effect, models that accurately
characterize overall mass loss rates during gasification and combustion at high
temperatures must be based on the intrinsic chemical reactivity of the carbonaceous
particle material, and parameters that control the mode of burning must depend on char
reactivity. Such models have been developed and partially validated in ongoing work in
our laboratory [3-6]. A variety of coals, biomass materials and synthetic chars have been
tested to determine parameters that describe char reactivity as a function of temperature
and reactive gas concentration. In a related study [7], a model for the mode of char
particle burning was validated for coal chars and the chars of biomass materials that are
relatively friable (and hence, can be pulverized). Almond shells and wood chips fall into
this category. Straws do not.
In our experimental approach, chars characteristic of those created at high
temperatures and heating rates in actual coal-fired boilers and furnaces are produced in a
laminar flow reactor. By properly adjusting the flow rates of the gases (CH4, H2, O2, and
N2) fed to the diffusion-flamelet burner of the flow reactor, gaseous environments can be
established inside the reactor that have a specified oxygen content (from trace amounts to
12-mol-% O2) at specified temperature (from 1300 to 2000 K). Size-classified samples
of the materials to be examined (coals, biomass materials, synthetic chars, and chars to be
heat-treated) are fed along the centerline of the flow reactor and partially reacted chars
are extracted at selected residence times (up to 200 ms) using a solids sampling probe.
The extracted samples are analyzed to determine char physical and chemical properties as
functions of char conversion. Particle size distributions are measured using a Coulter
Multiziser, an instrument that measures the size distributions of particle suspensions
using an electroresistive method. Apparent densities of particles are measured using a tap
density procedure in which packed-beds of particles in a graduated cylinder are weighed.
Specific surface areas of particles are determined from carbon dioxide adsorption
measurements employing CO2/He mixtures at 298 K and 10 atm. Intrinsic char
reactivities to oxygen are determined from oxidation tests performed in a pressurized
thermogravimetric analyzer (PTGA) under chemical kinetics-controlled reaction
conditions. Analytical procedures are discussed in various publications [3-7]. The data
are used to validate and to determine parameters in the sub-models developed.
Chars for testing are also produced in a tube furnace in nitrogen environments at
temperatures up to 1000 ºC at low heating rates (less than 50 ºC/min). These chars are
representative of those formed in grate furnaces, where particles experience a wide range
of heating rates in volatiles-rich environments. These chars are also subjected to
oxidation tests in the PTGA so that their intrinsic reactivities can be determined.
Figure 1 shows results of a typical oxidation test when wood chips are exposed to 10
mol-% O2 at 500 ºC. After the sample has been put in the PTGA balance pan and the

GCEP Technical Report - 2004 203

II.4 Project Results: Advanced Combustion

PTGA is closed, the reaction chamber is purged with nitrogen for 53 min at room
temperature to ensure an inert environment. Any CO2, CO or air, remaining in the
reaction chamber is eliminated during this time. After 53 min, the temperature is ramped
up to the reaction temperature of 500 ºC in 20 min, still in an oxygen-free environment in
order to dry the sample. As can be observed from the flat portion of the weight profile
from 75 to 90 min, the sample is dry before oxygen is admitted. During this heating and
drying of the sample in nitrogen, a small amount of CO and CO2 is released (desorbed).
These carbon oxides are formed from oxygen complexes in the initial char sample.

Nitrogen Reaction gas Nitrogen

25 1200
O2 Weight
Temp. CO

Temp. [ C], CO [ppm] and CO2

1000
m [mg] and O2 [x1.5 vol.%]

20 CO2
800

[x0.5 ppm]
15
600
10
400

o
5
200

0 0
0 100 200 300 400
Time [min.]

Figure 1: Experimental results from a typical oxidation test in 10 vol-% O2 at 1 atm.

At 90 min, the reaction gas (10 vol-% O2 in nitrogen) is turned on. From 90 to 330
min, the sample is kept at a constant temperature (500 ºC) and in a uniform and constant
gas composition. During the reaction period the oxygen reacts with the carbon in the
sample, producing CO and CO2, which are subsequently released, reducing the mass of
the material on the PTGA balance pan. At 330 min the reaction gas is turned off and the
sample is once again purged with nitrogen, making sure that the sample is no longer
reacting. At 378 min, still in nitrogen, the sample is heated up till 1100 ºC in order to
remove any adsorbed oxygen complexes remaining on the material in the balance pan.
This yields the true sample mass at the end of the test. The oxygen on the sample during
this desorption is released as CO and CO2 if there is still carbon left in the sample.
The mass profile between 90 and 330 min is differentiated to determine the overall
conversion rate in the environment established in the PTGA. The char particles are
assumed to consist of two components: an ash fraction, which is assumed to be non-
reactive, and a carbonaceous fraction, which is assumed to react, forming both CO and

204 GCEP Technical Report - 2004

 II.4 Project Results: Advanced Combustion

CO2. The specific mass loss rate of the char (Rc) is expressed in terms of the mass loss
rate and specific surface area of the carbonaceous portion of the particle material (Sgc) as:
1 dm c 1 dx c
Rc = − = = Ric Sgc , (1)
m c dt 1 − x c dt
where mc is the mass of carbonaceous material in the char at time t; xc is the fractional
char conversion, daf; Sgc is the specific surface area of the carbonaceous material; and Ric
is the intrinsic chemical reactivity per unit specific surface area.
The specific surface area model developed by Bhatia and Perlmutter [8] for porous
carbons is used to describe the variations in specific surface area of the carbonaceous
particle material with conversion under chemically-controlled oxidation conditions:
Sgc = Sgc,0 1− ψ ln(1− x c ). (2)

Here, xc is fractional conversion on a dry, ash-free basis (daf) and ψ is a structural

parameter that can be determined from fits to data. Our in situ surface area data obtained
with coal, biomass, and synthetic chars confirm the validity of this model for constant
apparent density burning, as is the case when rates of chemical reaction control overall
mass loss rates.
The specific surface area of an ash-laden char particle having an ash fraction Xash
after reaction for time t is assumed to obey the relation
Sgp = Xash Sga + (1− Xash )Sgc . (3)
Sgp is the quantity actually determined from the CO2-adsorption measurements. The
adsorption measurements taken at the beginning of an oxidation test yields Sgp,0, the
initial specific surface area of the char particle. Gas adsorption measurements taken at
the end of an oxidation test, after all the carbonaceous material has been burned away
yields Sga, the specific surface area of the ash. Equations (1) - (3) permit the
determination of char reactivity as a function of char conversion using the measured data.
Reactivities determined in PTGA environments of specified temperature and oxygen
content provide the data needed to gain an understanding of the important reaction
pathways governing char oxidation. In our approach, the following heterogeneous
reaction mechanism is used to characterize the reactivities:
2 Cf + O2 → 2 C(O) (R1)
Cf + C(O) + O2 → CO2 + C(O) + Cf (R2)
Cf + C(O) + O2 → CO + C(O) + C(O) (R3)
C(O) → CO + Cf (R4)
Here, Cf represents a free carbon site, one available for oxygen chemisorption, and C(O)
represents a carbon site filled with a chemisorbed O atom. It is assumed that each carbon
atom represents a potential adsorption site and that desorption of a chemisorbed oxygen
atom removes the associated carbon atom to uncover an underlying carbon site, which
becomes available for oxygen adsorption.

GCEP Technical Report - 2004 205

II.4 Project Results: Advanced Combustion

To date, Arrhenius parameters that describe the reaction rate coefficients for the
above reactions have been determined for several coal, biomass, and synthetic chars. All
of the chars examined were produced at high heating rates in the laminar flow reactor.
Using the kinetic parameters determined from analysis of the data obtained in the PTGA
oxidation tests, mass loss rates in the flow reactor can be predicted accurately using a
char combustion model that takes into account oxygen transport to the outer surfaces of
particles as well as oxygen diffusion through the pores of particles as the particles burn.
Recent Results
During the past year, we developed the capability to measure O2, CO, and CO2
concentrations in the reaction chamber of the PTGA during an oxidation test. The
plumbing and instrumentation needed to monitor the gases just above the sample in the
balance pan were installed and the software needed to integrate the signals from the gas
detectors with the weight measurements was written and compiled to run on the computer
that controls the PTGA. Calibration tests were performed as were validation tests with
empty balance pans and non-reactive, inert materials. The O2, CO, and CO2 profiles
shown in Fig. 1 are indicative of our present capability to make simultaneous gas and
weight loss measurements during an oxidation test. These measurements permit better
characterization of the rates of reactions R2 and R3 in the heterogeneous reaction
mechanism. They provide a direct measure of the heterogeneous CO-to-CO2 product
ratio as a function of temperature.
As part of our effort to assess the extent to which the reactivity of a biomass char can
be predicted based on its fractional contents of cellulose, hemicellulose and lignin, three
cellulosic-based biomass materials were tested to determine the reactivity of their chars to
oxygen. The materials selected (wood chips (WC), newspaper (NP), and glossy paper
(GP)) had similar chemical compositions on an ash-free basis, but differed in their ash-
content. Proximate and ultimate analyses are shown in Table I.
Table I: Proximate and ultimate analyses and specific surface areas of biomass materials
and their chars.
Proximate Analysis Ultimate Analysis (daf) Surface Area
VM Fix-C Ash C H Oa N S Sgp,0 Sgc,0
(wt%) (wt%) (wt%) (wt%) (wt%) (wt%) (wt%) (wt%) (m2/g) (m2/g)
Parent
WC 86.3 13.3 0.4 47.7 6.2 46.0 0.1 <0.02
NP 85.9 10.7 3.5 46.3 6.0 47.6 0.1 <0.02
GP 70.6 4.5 24.8 41.9 5.3 52.7 0.1 <0.02
Char
WC 7.6 90.2 2.2 98.58 0.4 0 1.0 0.02 389 398
NP 9.4 79.1 11.5 98.1 0.8 0 1.0 0.1 360 407
GP 21.4 16.8 61.8 73.6 1.7 23 1.1 0.6 140 366
a
Obtained by mass balance

Note the relatively high ash content of the glossy paper. It is likely that the clays and
other additives used as filling material to obtain a glossy finish contributed to the non-
combustible components of the paper. The compositions of the ashes were somewhat
similar, SiO2 and CaO accounting for about 60% of the weight of the ash of each parent

206 GCEP Technical Report - 2004

 II.4 Project Results: Advanced Combustion

material. About 35% of the weight of the ash of the glossy paper was unidentified, and is
believed to be associated with additives used in processing the paper.
Chars were produced by devolatilizing the biomass materials in nitrogen at low
heating rates in a tube furnace maintained at 900 ºC. The proximate and ultimate
analyses of the low-heating-rate chars are also shown in Table I. Note that the chars still
contain some volatile matter. The volatile matter content determined for the glossy paper
char is unusually high and so is the value reported for its oxygen content. Since the
oxygen content is determined by difference, it is likely that the value reported for the
oxygen weight fraction of the glossy paper char includes constituents of the additives
used in the finishing process. [The standard ultimate analysis procedure assumes that the
organic portion of the material examined only contains C, H, O, N, S, and Cl elements.]
Also presented in the table are the specific surface areas determined for the char particles.
Note that the specific surface areas of the carbonaceous portions of the chars are similar.
Intrinsic chemical reactivities determined for the chars of the three cellulosic-based
biomass materials when exposed to 10 mol-% oxygen at 500 ºC are presented in Fig. 2.
One of the newspaper chars was heat treated by injecting the char produced in the tube
furnace into the laminar flow reactor when an environment containing trace levels of
oxygen at 1550 K was established. This char is labeled NP-HT; its reactivity is also
shown in the figure.
3,6
3,2 WC-LH
2,8 NP-LH
NP-HT
Ri,c [10-6 gm-2s-1]

2,4
2 GP-LH

1,6
1,2
0,8
0,4
0
0 0,2 0,4 0,6 0,8 1
xc
Figure 2: Intrinsic reactivity (Ric) as a function of char conversion (xc).

The results reveal that the reactivities of the cellulosic-based chars selected for study
vary with conversion. All the chars have peak reactivity at low conversions (xc < 0.03,
daf). At the onset of oxidation, these low-heating-rate-produced chars exhibit a rapid
increase in reactivity followed by a rapid decrease. At ash-free conversions greater than
about 0.15, char reactivities are factors ranging from 3 to 6 below the peak values, and

GCEP Technical Report - 2004 207

II.4 Project Results: Advanced Combustion

remain at these relatively low levels until the final stages of burning. Note that subjecting
the newspaper char to a higher temperature had a relatively minor effect on its reactivity.
The heat-treated char does not, however, exhibit as sharp a peak as its parent char at the
onset of oxidation. It should be noted that all of the coal and biomass chars that we have
previously examined were produced at high heating rates and none of the chars exhibited
such sharp initial peaks in reactivity as observed with these chars that were produced
under low heating rate conditions.
The reactivity profiles of the chars produced at low heating rates are possibly the
consequence of the carbonaceous material consisting of two components, one that is
more reactive than the other. The initial peak in reactivity is due to the highly reactive
portion of the char, which is quickly consumed once the char is exposed to oxygen at
high temperatures. It may be that the more reactive portion of the carbonaceous material
is associated with the volatile matter content of the char. Low heating rates provide
opportunities for primary devolatilization products to undergo secondary reactions and
reattach to the carbonaceous matrix as they diffuse through devolatilizing particles. Such
reattached fragments are likely to be measured as volatile matter during a proximate
analysis. The glossy paper char has the highest peak reactivity; it also has the greatest
amount of volatile matter, as indicated by the proximate analysis. The peak reactivities
of the wood chip and newspaper chars are comparable; so are the volatile matter contents
of their chars. Additional tests are planned to determine if this is indeed the case. In
order to predict the types of reactivity profiles exhibited by these low-heating-rate chars,
it is necessary to modify our heterogeneous reaction mechanism to include two types of
carbon sites, one type being more reactive than the other. Such modifications are being
considered.
The data obtained with the glossy paper chars indicate a possible impact of the ash in
the char on inhibiting char oxidation rates near burnout. Note that the reactivity profile
for the glossy paper char falls to nearly zero at about 58% conversion, daf. The char
particles at this extent of conversion are ash-rich, containing more than 75% ash by
weight. It is likely that the ash plays a role in limiting the reactivity of the char by, for
example, encapsulating the carbonaceous material, rendering it more difficult for oxygen
to reach the carbonaceous surfaces where it can be adsorbed. By the time the wood chip
and newspaper chars are 75% ash by weight, char conversion is over 95%, daf. Any ash
inhibition effects are minor for these low-ash materials. Studies are underway to shed
insight into the possible inhibiting effect of ash, an important effect for biomass materials
having high ash contents.
Future Plans
Studies to determine the relationship between coal and biomass properties and model
parameters are ongoing as are studies to characterize the impact of the ash-content of
particles on char reactivity. A particular goal of our current research is to characterize the
impact of total pressure on coal- and biomass-char reactivity. To this end, coal and
biomass materials are being subjected to oxidation tests at high pressures. Our
heterogeneous reaction mechanism will be modified to reflect our latest understanding of
rates of the key chemical reactions controlling the conversion of the carbonaceous solid
material to gas-phase species. The objective of one of our studies is to assess the extent
to which the reactivity of a biomass char can be predicted based on its fractional contents

208 GCEP Technical Report - 2004

 II.4 Project Results: Advanced Combustion

of cellulose, hemicellulose, and lignin, the principle building blocks of biomass materials.
Towards meeting this goal, cellulose-rich, hemicellulose-rich, and lignin-rich biomass
materials are being identified and will be tested to determine the reactivity of their chars
to oxygen.
The work being performed will allow us to characterize accurately the chemical and
physical changes that coal and biomass particles undergo during combustion and
gasification processes. The studies undertaken will help us to understand how coal and
biomass properties influence char conversion rates in high-temperature, high-pressure
environments. The data obtained will permit the development and validation of the
physical and chemical sub-models needed in comprehensive models for coal-fired and
biomass-fired process units. The comprehensive models can be used to investigate
potential design strategies and can help define optimum operating conditions that yield
high coal and biomass conversion efficiencies with minimum impact on the environment.

References
1. http://www.fossil.energy.gov/programs/powersystems/combustion/
2. Handbook of Biomass Combustion and Co-Firing, S. van Loo and J. Koppejan, Eds.,
Twente University Press, The Netherlands, 2002.
3. Mitchell, R. E. An Intrinsic Kinetics-Based, Particle-Population Balance Model for
Char Oxidation During Pulverized Coal Combustion, Proc. Combust. Inst. 28, 2261-
2270, 2000.
4. Mitchell, R. E., Campbell, P. A., and Ma, L. Characterization of Coal-Char and
Biomass-Char Reactivities to Oxygen, Proc. Combust. Inst. 29, 519-526, 2002.
5. Mitchell, R. E. and Ma, L. Intrinsic Reactivity-Based Model for Mode of Particle
Burning, presented at the 3rd Joint Meeting of the U.S. Sections of the Combustion
Institute, Chicago, IL, March 16-19, 2003.
6. Mitchell, R. E. and Sørum, L. On the Reactivity of Chars from Cellulosic Wastes:
The Influence of Ash Content,” presented at the 3rd Joint Meeting of the U.S. Sections
of the Combustion Institute, Chicago, IL, March 16-19, 2003.
7. Mitchell, R. E. and Ma, L. A Mode of Burning Model During Oxidation of the Chars
of Pulverized Fuels: Model Implementation,” Paper 04S-45, Western States
Section/The Combustion Institute 2004 Spring Meeting, Davis, CA, March 29 - 30,
2004.
8. Bhatia, S. K., and Perlmutter, D. D. A Random Pore Model for Fluid-Solid
Reactions: I. Isothermal, Kinetic Control, AIChE Journal 26 (3) 379-386, 1980.

Contact
Reginald E. Mitchell: remitche@stanford.edu

GCEP Technical Report - 2004 209